The Effects of Light Intensity on the Flowering Phenology ...

Eastern Michigan University Eastern Michigan University

DigitalCommons@EMU DigitalCommons@EMU

Senior Honors Theses & Projects Honors College

2005

The effects of light intensity on the flowering phenology and inter-The effects of light intensity on the flowering phenology and inter-

population differences of the androdioecious plant species, population differences of the androdioecious plant species,

Mercurialis annua Mercurialis annua

Andrea Benedict

Follow this and additional works at: https://commons.emich.edu/honors

Recommended Citation Recommended Citation Benedict, Andrea, "The effects of light intensity on the flowering phenology and inter-population differences of the androdioecious plant species, Mercurialis annua" (2005). Senior Honors Theses & Projects. 74. https://commons.emich.edu/honors/74

This Open Access Senior Honors Thesis is brought to you for free and open access by the Honors College at DigitalCommons@EMU. It has been accepted for inclusion in Senior Honors Theses & Projects by an authorized administrator of DigitalCommons@EMU. For more information, please contact [email protected].

https://commons.emich.edu/

https://commons.emich.edu/honors

https://commons.emich.edu/honorscollege

https://commons.emich.edu/honors?utm_source=commons.emich.edu%2Fhonors%2F74&utm_medium=PDF&utm_campaign=PDFCoverPages

https://commons.emich.edu/honors/74?utm_source=commons.emich.edu%2Fhonors%2F74&utm_medium=PDF&utm_campaign=PDFCoverPages

mailto:[email protected]

The effects of light intensity on the flowering phenology and inter-population The effects of light intensity on the flowering phenology and inter-population differences of the androdioecious plant species, differences of the androdioecious plant species, Mercurialis annua Mercurialis annua

Abstract Abstract Mercurialis annua is an androdioecious plant species, meaning populations may contain both male and cosexual individuals. It is able to modify resource allocation to male and female function in varying conditions, such as light and competition regimes. One component of resource allocation may involve changes in the relative timing of production of male and female flowers. A controlled greenhouse experiment was performed to test for the effects of shading on the production of male and female flowers in four populations. Evidence suggested that plants grown in shade would be relatively less female than those grown in the sun. Although the average number of day to the production of male versus female flowers did not show a response to shading, the number of plants producing male versus female flowers during early development differed between light treatments. The shade treatment induced a reduction in number of plants producing female flowers, as predicted by my hypothesis. The populations differed in their specific responses to shading, suggesting that populations have adapted individually to local selective pressures.

Degree Type Degree Type Open Access Senior Honors Thesis

Department Department Biology

First Advisor First Advisor Dr. Gary Hannan

Second Advisor Second Advisor Dr. James Vandenbosch

Keywords Keywords Pollination, Plants Classification, Mercurialis Annua Speciation, Plant species

This open access senior honors thesis is available at DigitalCommons@EMU: https://commons.emich.edu/honors/74

https://commons.emich.edu/honors/74

https://commons.emich.edu/honors/74

Senior Honors Thesis

ABSTRACT

Author: Andrea L. Benedict

Department: Biology

Supervising Instructor: Dr. Gary Hannan

Honors Advisor: Dr. James Vandenbosch

Title: The Effects of Light Intensity on the Flowering Phenology and Inter-population

Differences of the Androdioecious Plant Species, Mercurialis annua

Length: 32 pages

Completion Date: 4-28-2005

________________________________________________________________________

Mercurialis annua is an androdioecious plant species, meaning populations may contain both

male and cosexual individuals. It is able to modify resource allocation to male and female

function in varying conditions, such as light and competition regimes. One component of

resource allocation may involve changes in the relative timing of production of male and female

flowers. A controlled greenhouse experiment was performed to test for the effects of shading on

the production of male and female flowers in four populations. Evidence suggested that plants

grown in shade would be relatively less female than those grown in the sun. Although the

average number of day to the production of male versus female flowers did not show a response

to shading, the number of plants producing male versus female flowers during early development

differed between light treatments. The shade treatment induced a reduction in number of plants

producing female flowers, as predicted by my hypothesis. The populations differed in their

specific responses to shading, suggesting that populations have adapted individually to local

selective pressures.

Table of Contents

Introduction..................................................................................................1

Methods........................................................................................................3

Results..........................................................................................................5

Discussion..................................................................................................25

Literature Cited ..........................................................................................30

Appendix A................................................................................................31

1

Introduction:

Mercurialis annua is a wind-pollinated, self-compatible, functionally

androdioecious plant species (Pannell 1997b). Androdioecy is a breeding system in

which there are both male and monoecious (cosexual) individuals coexisting in the same

population (Pannell 1997b). Androdioecy is a very rare breeding system and little is

known about the mechanisms by which males are maintained in a population (Barrett

2002).

All plants must partition the resources obtained from their environment between

growth and reproductive function (Obeso 2002). Cosexual plants, having both male and

female structures, must then partition the resources made available for reproduction in a

way that maximizes individual fitness (Parachnowitsch and Elle 2004). This partitioning

to maximize individual fitness is a key concept in sexual allocation theory

(Parachnowitsch and Elle 2004).

Female reproductive structures and the fruits they produce are more energetically

costly then male reproductive structures (Parachnowitsch and Elle 2004). The plants also

must dynamically allocate resources for reproduction at sequential nodes as the plants

increase in size (Kaitaniemi and Honkanen 1996). Mercurialis annua is unusual in being

androdioecious, where about 70% of plants in most populations are cosexual and about

30% are genetically male (Pannell, 1997a). Thus the sexual allocation theory may be

applied to the cosexual individuals in the population, but not the male plants, which are

only allocating resources to male reproductive function. Little is known about exactly

how androdioecy arose in this plant species, or why it may be advantageous to maintain

30% of the population as genetically male plants (Pannell 1997a). Evolutionary

2

pressures may cause cosexual plants to modify their allocation of resources to male

versus female reproductive function (Obeso 2002).

One very curious feature about Mercurialis annua is that, although maleness is

genetically determined by a single gene locus, some environmental factors, such as a

decrease in local competition, can cause some genetic males to function as cosexuals later

in their developmental process, and an increase in local competition can produce cosexes

with female-biased sex allocation (Pannell 1997c). Populations of Mercurialis annua in

the wild show great genetic reproductive diversity by being present in stands populations

of completely dioecious diploid individuals, and whereas other populations in other

stands, which are hexaploid and androdioecious (Pannell 1997b).

Another interesting attribute of this species is that the cosexual plants display

protogyny, which is the development of female reproductive organs before male organs, a

flowering behavior that to avoidreduces self-fertilization (Pannell 1997b). In a field

study, Pannell (1997a) found that shading by trees directly affected allocation for

reproduction; male plants reduced their allocation for pollen production, whereas

cosexuals reduced their allocation for female function, making the cosexuals

quantitatively more male. Since this was a field study it was not known whether some

other factor might be contributing to the shift towards maleness in cosexuals (Pannell

1997a).

An additional aspect of sexual reproduction in these plants may be in the timing

of the onset of female verses male reproduction. Previous work on sex allocation

showeds that Mercurialis annua is phenotypically plastic with respect to sexual function

(Pannell, 1997a,b,c). The objective of this experiment was to test the effects of light

3

intensity on flowering phenology in a greenhouse setting, while experimentally

controlling other factors that might influence flowering phenology.

My hypothesis is that if a reduction in female function occurs in shaded

conditions, then plants will either delay the onset of female flowering or fewer plants will

produce female flowers during the early stages of floral developmentfor which the plants

were observed. In addition, if populations are genetically variation in regardsvariable

with respect to control of their sexual function, then response to shading will vary

between populations.

Methods:

Mercurialis annua seeds were collected by Dr. John Pannell from two southern

Portugal populations (populations A and B) and two southern Spain populations

(populations C and D). A total of two hundred sixty-four264 2.5-inch pots were were

used. Sixty-six of the pots were used per population, initially with four seeds per pot,

planted in Fafard Superfine Germinating MixTM , on July 22, 2004. Thirty-three pots for

each population were randomly placed in either the sun or shade treatments. A total of

hundred thirty-two132 pots were placed in three randomly arranged sun treatment blocks

and one hundred thirty-two132 pots were placed in three randomly arranged shade tents.

Shade tents were constructed of PVC piping with two layers of burlap covering the top

and all four sides of the tent. The light level in the greenhouse on December 1, 2004 at

3:10 pm was measured at 125 µmol s-1 m-2, and the light level measured within the shade

tent was 38 µmol s-1 m-2, thus the tents provided 69.6% shade cover. After the plants had

germinated, nine days after planting, the pots were thinned down to one plant per pot,

with selection for plants with the tallest, darkest green and most robust leaves. Once

4

flowering commenced, the plants were checked regularly, about every two days. Nodes 0

through 5 of the plants were the only nodes observed in this experiment. The date of first

flowering and the sex of that flower at each node and at each position (axillary, stalk, and

axillary growth) at that node were recorded for each individual plant for each day data

was collectedof data collection (Figure 1).

Figure 1. Flowers of both sexes occurred at all three positions, and every node on the plant had

the possibility of producing all three positions at one time.

If the opposite sex came up at the same or at a later date at the same node and position

where a flower had already openedand/or already been recorded, the date was also

recorded for the opening of the opposite sex. Data collection for each of the light

treatments ended when more than half of the plants had flowered at the axillary position

at each of the first five nodes. The experiment was terminated on October 25, 2004.

DThe data analyses were performed on the mean number of days to first flowering using

a three-way analysis of variance (ANOVA) statistical test with light, node and sex as

factors using SPSS 11 for Macintosh. Since the number of plants producing flowers of

5

either sex was extremely variable and therefore, the ANOVA employed an unbalanced

design. Consequently, it was important to look at the number of plants producing flowers

of each sexual state. Pairwise , and aChi-Square tests were performed on the number of

plants producing flowers at each node in the sun and shade treatments to test for

independence between light, node, and sex for the number of plants producing flowers of

each sex..

Results:

Axillary Position: Number of Days to Flowering

The three-way ANOVA was used to test for significant effect of light, node, and

sex treatments and interactions among them. For all populations, the number of days to

first flowering differed between nodes (p<0.05) (Table 1). Most of the populations,

except population C, showed that the number of days to first flowering differed between

light treatments (p<0.05) (Table 1). There was no light by sex interaction for any of the

populations for the axillary position, which means that light treatment and sex

determination are independent of each other when determining the date of first flowering

(p>0.05). Populations A and C are showingshow more different combinations of

significant factors than the other populations and to each other. differences for which

factors are significant, as compared to pPopulations B and D, which are showng the same

factors as being significant (Table 1).

Table 1. Summary of ANOVA results testing for effects of light, node and sex on

number of days to first flowering.

Interaction

Population Light Node Sex Light X Node Node X Sex Light X Sex

A p=0.008 p=0.006 p=0.009 p=0.005 p=0.003 p=0.666

B p=0.034 p=0.013 p=0.148 p=0.804 p=0.252 p=0.732

6

C p=0.083 p=0.001 p=0.183 p=0.577 p=0.495 p=0.080

D p=0.026 p=0.004 p=0.144 p=0.308 p=0.505 p=0.813

Plants in the sun treatment always started producing flowers earlier than plants in

the shade treatment for all populations (Figure 2).

A B C D0

10

20

30

40

50

60

70

80

# of Days

Population

Sun

Shade

Figure 2. The mean number of days to first flowering is always shorter for the sun

treatment for all populations, as compared to the shade treatment.for sun and shade

treatments in the four populations. Bars represent + 1 standard error.

The node interaction effect can be attributed to the fact that flowers appeared

sequentially, starting with node one, for all of the populations (Figures 3-6). All

populations show female flowers occurring earlier than male flower at every node where

both sexes are present, except population D, node two (Figures 3-6). The ANOVA

results for populations B-D show no significant sex effect, but at most nodes female

flowers appear earlier than male flowers (Figures 4-6). Population A was the only

population to have a significant sex effect in the three-way ANOVA, meaning the day

7

that the first female flower matured was significantly earlier than the first maturing male

flower. Population A is the only population that showed both a ‘light by node’

interaction and a ‘node by sex’ interaction (Table 1). The ‘light by node’ interaction

shows that light treatment and node number influence each other with regard to the date

of first flowering. The ‘node by sex’ interaction shows that the node number and sex of

the flower influence each other in regards to the date of first flowering, which showed

that at node two, the female and male flowers appeared at about the same number of days

(Figure 3)

0

1

2

3

4

5

0 10 20 30 40 50 60 70 80 90

Nodes

# of Days

Female

Male

Figure 3. Number of days to maturation of the first flower of each sex at each node for

population A. Bars represent + 1 standard error.

8

0

1

2

3

4

5

0 10 20 30 40 50 60 70 80 90

Nodes

# of Days

Female

Male


population B. Bars represent + 1 standard error.

0

1

2

3

4

5

0 10 20 30 40 50 60 70 80 90

Nodes

# of Days

Female

Male


population C. Bars represent + 1 standard error.

9

0

1

2

3

4

5

0 10 20 30 40 50 60 70 80 90

Nodes

# of Days

Female

Male


population D. Bars represent + 1 standard error.

Axillary Position: Number of Plants Producing Flowers

Chi-Square tests were performed on the number of plants producing flowers to

look for significant interactions between light and sex, node and light, and node and sex.

Populations A and B showed a ‘light by sex’ interaction, but populations C and D did not

(Tables 2-5). Population A had more plants producing male flowers and fewer plants

producing male flowers than was expected in the sun treatment, and in the shade

treatment there were more plants producing female flowers than expected and fewer

plants producing male flowers (Table 2). Population B showed the opposite deviation

from expected numbers for the sun and shade treatments (Table 3). The sun treatment

had more plants producing female flowers than was expected and fewer plants producing

male flowers than was expected, and in the shade more plants produced male flowers and

fewer plants produced female flowers than was expected (Table 3). This interaction

occurreds when pooling all of the nodes togetherfrom each plant.

10

Table 2. Chi-Square test for independence between light and sex of flower with respect

to number of plants producing flowers of each sex for population A.

Sex

Light Female Male

Sun Observed 63.0 60.0

Expected 71.4 51.6

Shade Observed 60.0 29.0

Expected 51.6 37.4

X2 5.561

p-value 0.018


to number of plants producing flowers of each sex for population B.

Sex

Light Female Male


Expected 42.0 62.0


Expected 21.0 31.0

X2 7.668

p-value 0.006


to number of plants producing flowers of each sex for population C.

Sex

Light Female Male

Sun Observed 89 80

Expected 94.5 74.5

Shade Observed 71 46

Expected 65.5 51.5

X2 1.805

p-value 0.179

11


to number of plants producing flowers of each sex for population D.

Sex

Light Female Male

Sun Observed 66 89

Expected 68.8 86.3

Shade Observed 44 49

Expected 41.3 51.8

X2 0.527

p-value 0.468

Populations A and C showed a significant node by light interaction for the

number of plants producing flowers, but in populations B and D this interaction was not

significant (Tables 6-9). All four populations show the same pattern in the sun by having

more plants producing flowers than expected at the lower nodes and then having fewer

plants than expected flowering at the upper nodes. The shade treatment showed the

opposite effect with fewer plants producing flowers than were expected at the lower

nodes and more plants than expected producing flowers at the upper nodes. While this

general pattern was seen for all of the populations, only populations A and C deviated

enough from the expected numbers to provide significant Chi-Square values.

12

Table 6. Chi-Square test for independence between node and light treatment with respect

to number of plants producing flowers for population A.

Light

Node Sun Shade

5 Observed 24.0 25.0

Expected 28.4 20.6


Expected 30.2 21.8


Expected 34.8 25.2


Expected 22.6 16.4

1 Observed 12.0 0.0

Expected 7.0 5.0

0 Observed N/A N/A

Expected N/A N/A

X2 13.442

p-value 0.009


to number of plants producing flowers for population B.

Light

Node Sun Shade


Expected 25.3 12.7


Expected 32.7 16.3


Expected 26.7 13.3

2 Observed 20.0 4.0

Expected 16.0 8.0

1 Observed 5.0 0.0

Expected 3.3 1.7

0 Observed N/A N/A

Expected N/A N/A

X2 7.121

p-value 0.130

13


to number of plants producing flowers for population C.

Light

Node Sun Shade


Expected 37.2 25.8


Expected 43.7 30.3


Expected 44.3 30.7


Expected 35.5 24.5

1 Observed 14.0 0.0

Expected 8.3 5.7

0 Observed N/A N/A

Expected N/A N/A

X2 11.535

p-value 0.021


to number of plants producing flowers for population D.

Light

Node Sun Shade


Expected 39.4 23.6


Expected 46.3 27.8


Expected 37.5 22.5


Expected 28.1 16.9

1 Observed 6.0 0.0

Expected 3.8 2.3

0 Observed N/A N/A

Expected N/A N/A

X2 9.172

p-value 0.057

The ‘node by light’ interaction was further investigated by analyzing sexes separately for

each light treatment. All of the populations showed a ‘node by sex’ interaction with the

14

sun treatment for the number of plants producing flowers of each sex, but this interaction

was not significant in the shade treatment for any of the populations (Tables 10-13). In

the sun treatment the number of plants producing female flowers started off fewer plants

than the expected number at the lower nodes, and becoming more than expected at the

upper nodes. Whereas the number of plants producing male flowers showed the opposite

effect. The lower nodes had more plants producing male flowers than expected, and at

the upper nodes fewer plants than expected were producing male flowers. This deviation

from expected values was not seen in the shade treatment.

Table 10. Chi-Square test for independence between node and sex of flower with respect

to number of plants producing flowers of each sex in the sun and shade treatment for

population A.

Sun Shade

Sex Sex

Node Female Male Node Female Male

5 Observed 22.0 2.0 5 Observed 21.0 4.0

Expected 12.3 11.7 Expected 16.9 8.1







1 Observed 0.0 12.0 1 Observed N/A N/A

Expected 6.1 5.9 Expected N/A N/A

0 Observed N/A N/A 0 Observed N/A N/A

Expected N/A N/A Expected N/A N/A

X2 39.339 X

2 5.608

p-value 0.000 p-value 0.132

15



population B.

Sun Shade

Sex Sex














X2 17.323 X

2 3.788




population C.

Sun Shade

Sex Sex














X2 32.551 X

2 0.386


16



population D.

Sun Shade

Sex Sex














X2 19.105 X

2 1.543


Stalk Position: Number of Days to First Flowering Results

ANOVA was not performed on stalk data because during the course to the

experiment few plants produced flowers at this position. The small sample sizes and lack

of female flowers in general made ANOVA tests impractical. General patterns can be

interpreted from graphs that show the number of days to maturation of the first flower of

each sex at each node for each of the populations (Figure 7-10). Population A showed

that nodes flowered sequentially starting with the lower nodes and that the stalks

produced male flowers earlier than female flowers when both sexes appeared at a given

node (Figure 7). In population B, nodes flowered sequentially, starting with the lower

nodes, and the stalks produced female flowers earlier than male flowers when both sexes

appeared at a given node (Figure 8). In population C nodes flowered sequentially,

starting with the lower nodes; female flowers matured earlier than male flowers when

both sexes appeared at a given node, except at nodes one and two where the male flowers

17

matured slightly before the females (Figure 9). Population D showed that nodes started

flowering sequentially, starting with the lower nodes, and that the stalks produced almost

exclusively male flowers, except for one female flower at node four (Figure 10).

0

1

2

3

4

5

0 10 20 30 40 50 60 70 80 90

Nodes

# of Days

Female

Male



18

0

1

2

3

4

5

0 10 20 30 40 50 60 70 80 90

Nodes

# of Days

Female

Male



0

1

2

3

4

5

0 10 20 30 40 50 60 70 80 90

Nodes

# of Days

Female

Male



19

0

1

2

3

4

5

0 10 20 30 40 50 60 70 80 90

Nodes

# of Days

Female

Male



Stalk Position: Number of Plants Producing Flowers

Chi-Square tests were performed on the number of plants producing flowers to

look for significant interactions between light and sex. Flowering at individual nodes

was not analyzed because few plants produced flowers on stalks at any given node.

Populations A and C showed a significant light by sex interaction (Tables 14 and 16).

For both populations, the number of plants producing female flowers was lower than

expected in the sun, and higher than expected in the shade (Tables 14 and 16). Both

populations also showed that the number of plants producing male flowers was higher

than expected in the sun and lower than expected in the shade, opposite from the trend

seen for the number of plants producing female flowers for both populations (Tables 14

and 16). Populations B and D did not show a significant ‘light by sex’ interaction.

(Tables 15 and 17). Population C had many more plants producing female flowers on

20

stalks (Table 16) than the other three populations, which had very few (Tables 14, 15, and

17).


to number of plants producing flowers of each sex for population A.

Sex

Light Female Male


Expected 5.0 53.0


Expected 2.0 21.0

X2 6.979

p-value 0.008


to number of plants producing flowers of each sex for population B.

Sex

Light Female Male


Expected 5.4 43.6


Expected 5.6 44.4

X2 0.990

p-value 0.320


to number of plants producing flowers of each sex for population C.

Sex

Light Female Male


Expected 17.6 26.4


Expected 12.4 18.6

X2 7.185

p-value 0.007

21


to number of plants producing flowers of each sex for population D.

Sex

Light Female Male


Expected 0.6 26.4


Expected 0.4 18.6

X2 1.453

p-value 0.228

Axillary Growth Position: Number of Days to Flowering

ANOVA was not performed on the axillary growth data because during the course

of the experiment, few plants produced flowers at this position. The small sample sizes

and lack of female flowers in general precluded the use of ANOVA tests. General

patterns can be interpreted from graphs that show the number of days to maturation of the

first flower of each sex at each node for each of the populations (Figure 11-14).

Population A started producing female flowers at sequential nodes, starting at node two

(Figure 11). There were only two plants that produced male flowers in population A

(Figure 11). Population B did not show any type of consistent pattern in the number of

days to maturation of both sexes (Figure12). In population C, nodes started flowering

sequentially starting with the node one, but node zero deviated from this pattern by

producing mature flowers later than node one (Figure 13). Population C also showed that

female flowers matured first, except at node two (Figure 13). Population D showed that

the nodes started flowering sequentially starting with the node one, and the flowers are

almost all female except at node one (Figure 14). Very few plants in the four populations

produced flowers at the axillary growth position. In some cases only a single flowering

event was observed for a given node.

22

0

1

2

3

4

5

0 10 20 30 40 50 60 70 80 90

Nodes

# of Days

Female

Male



0

1

2

3

4

5

0 10 20 30 40 50 60 70 80 90 100

Nodes

# of Days

Female

Male



23

0

1

2

3

4

5

0 10 20 30 40 50 60 70 80 90 100

Nodes

# of Days

Female

Male



0

1

2

3

4

5

0 10 20 30 40 50 60 70 80 90

Nodes

# of Days

Female

Male



24

Axillary Growth Position: Number of Plants Producing Flowers

Chi-Square tests were not performed on the axillary growth data because very few

plants produced flowers at the axillary growth position making Chi-Square tests

undependable. However, a comparison of the observed data and expected values for a

‘light by sex’ cross-tabulation shows that populations A, B, and D closely match

expected values for the number of plants producing each sex of flower in each light

treatment (Tables 18, 19, and 21). Population C shows deviation from the expected

numbers in that in the sun there were fewer plants producing female flowers than

expected and more producing males than expected and in the shade there were more

plants producing female flowers than expected and fewer producing males (Table 20).

Population C had many more plants producing female flowers at the axillary growth

position (Table 20) than the other three populations which had very few (Tables 18, 19,

and 21).

Table 18. Crosstabs results for number of plants producing each sex of flowers for both

light treatments for population A.

Sex

Light Female Male


Expected 7.4 1.6


Expected 6.6 1.4


light treatments for population B.

Sex

Light Female Male


Expected 4.0 2.0


Expected 2.0 1.0

25


light treatments for population C.

Sex

Light Female Male


Expected 62.3 17.7


Expected 18.7 5.3


light treatments for population D.

Sex

Light Female Male


Expected 6.4 0.6


Expected 3.6 0.4

Discussion:

ANOVA results for the number of days to first flowering for the axillary position

showed that plants behaved differently in each light treatment. The plants in the sun

consistently matured earlier than the plants in the shade. The shaded plants were very

lanky in appearance compared to the sun plants, and they were later in germination time,

growth, and reaching the point of sexual maturity. The plants in both sun and shade were

also flowering at sequential nodes in the axillary, stalk, and axillary growth positions

starting at the lower nodes. In regards to which sex of flower matured first in each light

treatment, female flowers appeared first in both the sun and shade treatments and for all

of the positions. However, both the stalk and axillary growth position produced very few

flowers in general and were male flower biased and female flower biased, respectively.

The lack of a ‘light by sex’ interaction showed that the plants were not changing the

pattern for when each sex of flower first matured in the different light conditions; the

26

plants in the shade exhibited similar delays in maturity for both male and female flowers.

Overall, the number of days to flowering results did not support the hypothesis that plants

in the shade treatment will delay the onset of female flowers as an indicator of decreased

allocation to female function, an hypothesis based on Pannell’s (1997a) field

observations. Although this hypothesis was not supported, the fact that the populations

differed in which factors influenced the number of days to first flowering suggests that

populations differ in flowering behavior. Populations may have evolved different

evolutionary responses to shading, an environmental stress.

It was evident when observing the plants that not all plants produced flowers of

both sexes at all nodes and at all positions. This lack of flowering at all nodes and

position combinations may have been a consequence of arbitrarily limiting the

observation period. Consequently, the plants had not reached full sexual maturity by the

end of the experiment. For some of the node and position combinations only one plant in

the whole population produced a flower. Thus, the mean number of days to first

flowering may have been greatly influenced by small sample sizes for the populations.

Analyzing the number of plants producing flowers may be a better indicator of how the

plants are behaving under varying light conditions.

Chi-Square tests of independence between light conditions and number of plants

producing flowers of each sex, lumping all nodes together, showed that some populations

at each position exhibited a significant interaction between light and sex. The stalk and

axillary growth positions were not analyzed by node because there were not enough

plants producing flowers of either sex at these positions to be able to analyze individual

nodes. Within the populations that showed significant interactions at each position, the

27

contrasting light treatments differed in whether male or female flowers were more

numerous than expected. For population A (axillary position), populations A and C

(stalk position), and population C (axillary growth position), there were fewer than the

expected number of plants producing female flowers and more than the expected number

of plants producing male flowers in the sun treatment. In the shade treatment, a larger

number of plants than expected produced female flowers.

When looking for significance in the node by light interaction in the axillary

position, the nodes were separated and the sexes were lumped together. All of the

populations showed the same pattern, with the sun treatment having more plants

producing flowers at the lower nodes than was expected and having fewer plants

producing flowers at the upper nodes than was expected. The shade treatment showed

the opposite effect and had fewer plants producing flowers at the lower nodes and had

more plants producing flowers at the upper nodes than expected. Only populations A and

C actually deviated enough from the expected values to provide significant Chi-Square

values. Since this was a consistent pattern seen when the nodes were separated, but the

sexes were lumped together, it was important to look at how each of the sexes were

influencing these results.

For the ‘node by sex’ interaction, where the light treatments analyzed separately,

all of the populations showed a consistent pattern. In the sun treatment there were fewer

plants producing female flowers at the lower nodes and more plants producing female

flowers at the upper nodes than was expected. The number of plants producing male

flowers showed the opposite pattern by having more than expected at lower nodes and

fewer than expected at the upper nodes. This pattern showed a shift towards more plants

28

producing female flowers at the upper nodes as the plant matured, thus becoming more

female. This shift towards femaleness at later nodes was not seen in the shade treatment.

The number of plants producing each sex of flower in the shade does not show any sort

of nodal pattern. These results show that lumping the sexes together for the ‘node by

light’ interaction test swamped out the individual pattern of flower production by sex in

the sun treatment.

The results of the ‘node by sex’ Chi-Square test support the hypothesis that, in the

shade, fewer plants will produce female flowers, taking resources away from female

function. The fact that there was no shift towards femaleness at the upper nodes in the

shade shows that the plants were behaving differently under this environmental stress.

The interaction between light and sex in all three populations produced a reversal in

which of the sexes was more frequent than expected in the shade as compared to the sun.

The observation that population B (axillary position) differs from the other populations

and positions by being female biased in the sun treatment suggests that populations may

have evolved different allocation responses to changes in light conditions.

An example of how the populations may be evolving different sex allocation

responses to environmental stress may be illustrated by behavior of plants in population

C. Plants in this population were more highly branched and had more axillary

reproductive structures than did plants in the other populations. This population also had

more plants producing flowers of either sex at all of the growth positions.

The fact that female flowers appear first overall helps to explain why the plants

shift their allocation away from female function in the shade. Plants in the shaded

conditions have more limited resources than plants in the sun, and thus have fewer

29

resources to allocate to reproductive function. Since female flowers and the fruits they

produce are more expensive then male flowers, and in this plant species females mature

first, the greatest impact on resource use by the plants should occur by shifting resources

away from the more energetically costly female flowers.

The arbitrary cut-off date may have unintentionally swayed the results because

there was no real way of knowing if the plants would have shown different patterns if

plants had been followed to maturity. Time constraints on this project did not allow for

further observations past the arbitrary cut off date. The prior study looking at the effects

of shading on Mercurialis annua used seed count, dry fruit weight per plant biomass, and

pollen production as variables to test for shading effects, which may be better indicators

of sex allocation in this species than the number of days to flowering or the number of

plants producing each sex of flowers (Pannell, 1997a). The previous study also harvested

the plants for data collection, but this study allowed the plants to be monitored over a

portion of their life cycle, producing different types of result than previously reported

(Pannell 1997a).

Overall, this study helps to support the field observations that cosexual plants

from androdioecious Mercurialis annua populations reduce their allocation to female

function in the shade as compared to those growing in the full sun (Pannell 1997a). The

four populations studied also were responding to shading differently and thus provide an

opportunity for the evolution of different resource allocation patterns in this species. This

is one of the few studies to examine the role of environmental stresses on the phenology

of flower production in an androdioecious species.

30

Literature Cited:

Barrett, S. 2002. The evolution of plant sexual diversity. Nature Reviews Genetics

3: 274-374.

Kaitaniemi, P. and Honkanen T. 1996. Simulating source-sink control and nutrient

translocation in a modular plant. Ecological Modelling. 88:227-240.

Obeso, J. R. 2002. Tansley review no.139: the costs of reproduction in plants. New

Phytologist. 155: 321-348.

Pannell, J. 1997a. Variation in sex ratios and sex allocation in androdioecious

Mercurialis annua. Journal of Ecology 85: 57-69.

Pannell, J. 1997b. Widespread functional androdioecy in Mercurialis annua L.

(Euphorbiaceae). Biological Journal of the Linnean Society 61: 95-116.

Pannell, J. 1997c. Mixed genetic and environmental sex determination in an

androdioecious population of Mercurialis annua. Heredity 78: 50-56.

Parachnowitsch, A.L., and Elle, E. 2004. Variation in sex allocation and male-female

trade-offs in six populations of Collinsia parviflora (Scrophulariaceae s.l.).

American Journal of Botany. 91: 1200-1207.

31

Appendix A. Means, standard errors, and sample sizes for number of days to flowering

in axillary position for populations A-D, broken down by light, sex and node.

POP LIGHT SEX NODE N Mean Std. Error of Mean

2 7 33.29 4.455 3 14 35.36 1.041

4 20 46.30 2.139

female

5 22 58.64 1.652

male 1 12 33.58 1.390

2 20 38.45 .966

3 17 44.47 1.065

4 9 57.00 3.346

sun

5 2 63.00 13.000

shade 2 7 46.57 1.251

3 16 54.13 2.674 4 16 63.50 2.703

female

5 21 76.76 1.600

male 2 5 52.80 2.888

3 13 63.08 1.838

4 7 74.14 2.857

A

5 4 82.00 3.697

B sun 2 5 32.20 1.934

3 11 39.09 2.011

4 17 48.12 1.946

female

5 17 62.94 1.158

male 1 5 34.20 1.744 2 15 39.53 .910

3 15 48.53 1.897

4 14 59.93 2.655

5 5 71.40 2.421

shade 2 1 54.00 .

3 1 54.00 .

4 5 55.60 1.470

female

5 6 73.83 3.371

male 2 3 50.33 2.028

3 13 60.00 1.905 4 13 74.23 2.972

5 10 84.40 3.334

32

POP LIGHT SEX NODE N Mean Std. Error of Mean

C sun 2 14 28.07 1.131

3 22 35.14 1.037

4 25 41.88 1.531

female

5 28 53.46 1.959

male 1 14 33.00 1.011

2 24 37.67 .906

3 20 41.75 1.138

4 16 49.31 2.213

5 6 55.50 4.631

shade 2 14 46.86 1.382

3 19 52.47 2.269

4 21 59.33 1.764

female

5 17 69.71 1.504

male 2 8 61.63 4.807

3 14 60.93 1.542

4 12 69.50 2.024

5 12 83.17 2.153

D sun 1 1 33.00 .

2 7 33.71 2.437

3 14 39.57 1.401

4 18 50.78 2.907

female

5 26 59.69 1.747

1 5 35.80 .800

2 26 39.23 1.011

3 25 46.00 1.361

4 21 57.14 2.242

male

5 12 67.17 3.563

2 6 52.67 7.504

3 8 53.63 5.165

4 16 55.44 2.143

female

5 14 63.64 2.645

male 2 6 50.17 .749

3 13 60.69 1.998

4 19 71.11 2.321

shade

5 11 72.73 2.711

Date post:	05-Dec-2021
Category:	Documents
Upload:	others
View:	2 times
Download:	0 times

The Effects of Light Intensity on the Flowering Phenology ...

Documents